Re: [FFmpeg-devel] [PATCH 2/2] avcodec: remove anatoliy prores encoder

[Paul B Mahol]

On 4/26/18, Carl Eugen Hoyos  wrote:
> 2018-04-26 13:48 GMT+02:00, Paul B Mahol :
>> On 4/26/18, Carl Eugen Hoyos  wrote:
>>> 2018-04-26 13:34 GMT+02:00, Paul B Mahol :
 On 4/26/18, Carl Eugen Hoyos  wrote:
> 2018-04-26 13:17 GMT+02:00, Josh de Kock :
>> On 2017/06/26 15:09, Paul B Mahol wrote:
>>> Rationale:
>>> - Slower then other encoder
>>> - Less configurable
>>> - Does not support alpha profile
>>> - Does not set interlaced flag
>>> - Worse output quality
>>> - No need for 2 encoders
>>>
>>> Signed-off-by: Paul B Mahol 
>>>
>> Is there any reason this was not pushed?
>> I can't seem to see any argument against it.
>
> It was shown in the past that this encoder is faster,
> more efficient and produces better quality.

 Why are you not telling real truth?
>>>
>>> This is surprisingly rude:
>>> I am always trying to tell the truth, one of the things
>>> that make me less happy about contributing here is
>>> both that I am not allowed to write the truth anymore.
>>>
>>> Anyway: In the discussion about adding one of the
>>> features you mention above, tests were posted that
>>> showed this encoder to produce better (objective,
>>> with all its disadvantages) quality using measurably
>>> less cycles.
>>
>> That was with default configuration for both of them.
>
> Which is not the most likely usecase?

Even if it is, that is unfair comparison.
Because options and settings are incompatible.

I hope you understand this.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] [PATCH 2/2] avcodec: remove anatoliy prores encoder

[Nicolas George]

Paul B Mahol (2018-04-26):
> That was with default configuration for both of them.

Then please post a patch to enhance the default configuration.

Regards,

-- 
  Nicolas George


signature.asc
Description: Digital signature
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] [PATCH 2/2] avcodec: remove anatoliy prores encoder

[Carl Eugen Hoyos]

2018-04-26 13:48 GMT+02:00, Paul B Mahol :
> On 4/26/18, Carl Eugen Hoyos  wrote:
>> 2018-04-26 13:34 GMT+02:00, Paul B Mahol :
>>> On 4/26/18, Carl Eugen Hoyos  wrote:
 2018-04-26 13:17 GMT+02:00, Josh de Kock :
> On 2017/06/26 15:09, Paul B Mahol wrote:
>> Rationale:
>> - Slower then other encoder
>> - Less configurable
>> - Does not support alpha profile
>> - Does not set interlaced flag
>> - Worse output quality
>> - No need for 2 encoders
>>
>> Signed-off-by: Paul B Mahol 
>>
> Is there any reason this was not pushed?
> I can't seem to see any argument against it.

 It was shown in the past that this encoder is faster,
 more efficient and produces better quality.
>>>
>>> Why are you not telling real truth?
>>
>> This is surprisingly rude:
>> I am always trying to tell the truth, one of the things
>> that make me less happy about contributing here is
>> both that I am not allowed to write the truth anymore.
>>
>> Anyway: In the discussion about adding one of the
>> features you mention above, tests were posted that
>> showed this encoder to produce better (objective,
>> with all its disadvantages) quality using measurably
>> less cycles.
>
> That was with default configuration for both of them.

Which is not the most likely usecase?

> That's like comparing apples and oranges.
>
> I fail to see how that makes your statements true.

You seem to remember the situation even better
than I do, so I really don't understand your
comments...

>>> None of your claims are really true.
>>
>> Given how much you embarrassed us in the prores
>> discussion, I wonder why you make this claim;-)
>
> Now, this is just rude.

No?

> I expected this all the time.

You expected all the time that prores actually is 12bit?
(I didn't)

Carl Eugen
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] [PATCH 2/2] avcodec: remove anatoliy prores encoder

[wm4]

On Thu, 26 Apr 2018 13:39:56 +0200
Carl Eugen Hoyos  wrote:

> 2018-04-26 13:34 GMT+02:00, Paul B Mahol :
> > On 4/26/18, Carl Eugen Hoyos  wrote:  
> >> 2018-04-26 13:17 GMT+02:00, Josh de Kock :  
> >>> On 2017/06/26 15:09, Paul B Mahol wrote:  
>  Rationale:
>  - Slower then other encoder
>  - Less configurable
>  - Does not support alpha profile
>  - Does not set interlaced flag
>  - Worse output quality
>  - No need for 2 encoders
> 
>  Signed-off-by: Paul B Mahol 
>   
> >>> Is there any reason this was not pushed?
> >>> I can't seem to see any argument against it.  
> >>
> >> It was shown in the past that this encoder is faster,
> >> more efficient and produces better quality.  
> >
> > Why are you not telling real truth?  
> 
> This is surprisingly rude:
> I am always trying to tell the truth, one of the things
> that make me less happy about contributing here is
> both that I am not allowed to write the truth anymore.

What would those truths be? Note: insults are not truths.

> Anyway: In the discussion about adding one of the
> features you mention above, tests were posted that
> showed this encoder to produce better (objective,
> with all its disadvantages) quality using measurably
> less cycles.
> 
> > None of your claims are really true.  
> 
> Given how much you embarrassed us in the prores
> discussion, I wonder why you make this claim;-)

That is surprisingly rude.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] [PATCH 2/2] avcodec: remove anatoliy prores encoder

[Paul B Mahol]

On 4/26/18, Carl Eugen Hoyos  wrote:
> 2018-04-26 13:34 GMT+02:00, Paul B Mahol :
>> On 4/26/18, Carl Eugen Hoyos  wrote:
>>> 2018-04-26 13:17 GMT+02:00, Josh de Kock :
 On 2017/06/26 15:09, Paul B Mahol wrote:
> Rationale:
> - Slower then other encoder
> - Less configurable
> - Does not support alpha profile
> - Does not set interlaced flag
> - Worse output quality
> - No need for 2 encoders
>
> Signed-off-by: Paul B Mahol 
>
 Is there any reason this was not pushed?
 I can't seem to see any argument against it.
>>>
>>> It was shown in the past that this encoder is faster,
>>> more efficient and produces better quality.
>>
>> Why are you not telling real truth?
>
> This is surprisingly rude:
> I am always trying to tell the truth, one of the things
> that make me less happy about contributing here is
> both that I am not allowed to write the truth anymore.
>
> Anyway: In the discussion about adding one of the
> features you mention above, tests were posted that
> showed this encoder to produce better (objective,
> with all its disadvantages) quality using measurably
> less cycles.

That was with default configuration for both of them.
That's like comparing apples and oranges.

I fail to see how that makes your statements true.

>
>> None of your claims are really true.
>
> Given how much you embarrassed us in the prores
> discussion, I wonder why you make this claim;-)

Now, this is just rude. I expected this all the time.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] [PATCH 2/2] avcodec: remove anatoliy prores encoder

[Carl Eugen Hoyos]

2018-04-26 13:34 GMT+02:00, Paul B Mahol :
> On 4/26/18, Carl Eugen Hoyos  wrote:
>> 2018-04-26 13:17 GMT+02:00, Josh de Kock :
>>> On 2017/06/26 15:09, Paul B Mahol wrote:
 Rationale:
 - Slower then other encoder
 - Less configurable
 - Does not support alpha profile
 - Does not set interlaced flag
 - Worse output quality
 - No need for 2 encoders

 Signed-off-by: Paul B Mahol 

>>> Is there any reason this was not pushed?
>>> I can't seem to see any argument against it.
>>
>> It was shown in the past that this encoder is faster,
>> more efficient and produces better quality.
>
> Why are you not telling real truth?

This is surprisingly rude:
I am always trying to tell the truth, one of the things
that make me less happy about contributing here is
both that I am not allowed to write the truth anymore.

Anyway: In the discussion about adding one of the
features you mention above, tests were posted that
showed this encoder to produce better (objective,
with all its disadvantages) quality using measurably
less cycles.

> None of your claims are really true.

Given how much you embarrassed us in the prores
discussion, I wonder why you make this claim;-)

Carl Eugen
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] [PATCH 2/2] avcodec: remove anatoliy prores encoder

[wm4]

On Thu, 26 Apr 2018 13:34:12 +0200
Paul B Mahol  wrote:

> On 4/26/18, Carl Eugen Hoyos  wrote:
> > 2018-04-26 13:17 GMT+02:00, Josh de Kock :  
> >> On 2017/06/26 15:09, Paul B Mahol wrote:  
> >>> Rationale:
> >>> - Slower then other encoder
> >>> - Less configurable
> >>> - Does not support alpha profile
> >>> - Does not set interlaced flag
> >>> - Worse output quality
> >>> - No need for 2 encoders
> >>>
> >>> Signed-off-by: Paul B Mahol 
> >>>  
> >> Is there any reason this was not pushed?
> >> I can't seem to see any argument against it.  
> >
> > It was shown in the past that this encoder is faster,
> > more efficient and produces better quality.  
> 
> Why are you not telling real truth?
> 
> None of your claims are really true.
> 
> I wonder why.

Can any of you post numbers?
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] [PATCH 2/2] avcodec: remove anatoliy prores encoder

[Paul B Mahol]

On 4/26/18, Carl Eugen Hoyos  wrote:
> 2018-04-26 13:17 GMT+02:00, Josh de Kock :
>> On 2017/06/26 15:09, Paul B Mahol wrote:
>>> Rationale:
>>> - Slower then other encoder
>>> - Less configurable
>>> - Does not support alpha profile
>>> - Does not set interlaced flag
>>> - Worse output quality
>>> - No need for 2 encoders
>>>
>>> Signed-off-by: Paul B Mahol 
>>>
>> Is there any reason this was not pushed?
>> I can't seem to see any argument against it.
>
> It was shown in the past that this encoder is faster,
> more efficient and produces better quality.

Why are you not telling real truth?

None of your claims are really true.

I wonder why.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] [PATCH 2/2] avcodec: remove anatoliy prores encoder

[Josh de Kock]

On 2018/04/26 12:26, Carl Eugen Hoyos wrote:

2018-04-26 13:17 GMT+02:00, Josh de Kock :

On 2017/06/26 15:09, Paul B Mahol wrote:

Rationale:
- Slower then other encoder
- Less configurable
- Does not support alpha profile
- Does not set interlaced flag
- Worse output quality
- No need for 2 encoders

Signed-off-by: Paul B Mahol 


Is there any reason this was not pushed?
I can't seem to see any argument against it.


It was shown in the past that this encoder is faster,
more efficient and produces better quality.

I don't know when this has changed.



Ok, there was some discussion on IRC, so I was unsure. I will look into 
it again.


--
Josh
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] [PATCH 2/2] avcodec: remove anatoliy prores encoder

[Carl Eugen Hoyos]

2018-04-26 13:17 GMT+02:00, Josh de Kock :
> On 2017/06/26 15:09, Paul B Mahol wrote:
>> Rationale:
>> - Slower then other encoder
>> - Less configurable
>> - Does not support alpha profile
>> - Does not set interlaced flag
>> - Worse output quality
>> - No need for 2 encoders
>>
>> Signed-off-by: Paul B Mahol 
>>
> Is there any reason this was not pushed?
> I can't seem to see any argument against it.

It was shown in the past that this encoder is faster,
more efficient and produces better quality.

I don't know when this has changed.

Carl Eugen
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] [PATCH 2/2] avcodec: remove anatoliy prores encoder

[Josh de Kock]

On 2017/06/26 15:09, Paul B Mahol wrote:

Rationale:
- Slower then other encoder
- Less configurable
- Does not support alpha profile
- Does not set interlaced flag
- Worse output quality
- No need for 2 encoders

Signed-off-by: Paul B Mahol 

Is there any reason this was not pushed? I can't seem to see any 
argument against it.


--
Josh
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] [PATCH 2/2] avcodec: remove anatoliy prores encoder

[Jan Ekstrom]

On Mon, Jun 26, 2017 at 5:09 PM, Paul B Mahol  wrote:
> Rationale:
> - Slower then other encoder
> - Less configurable
> - Does not support alpha profile
> - Does not set interlaced flag
> - Worse output quality
> - No need for 2 encoders
>

I agree with this clean-up. Probably needs an API/ABI bump?

Jan
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel